You are here

Article Text

Article menu

Download PDFPDF

Original research
Radiologic Assessment of Groin Lymph Nodes in Pelvic Malignancies
  1. Soumon Rudra1,
  2. Dominique Fuser2,
  3. Todd A DeWees3,
  4. Leping Wan1,
  5. Margery Gang1,
  6. Caressa Y Hui1,
  7. Yuan J Rao4,
  8. Barry A Siegel2,5,
  9. Farrokh Dehdashti2,5,
  10. David G Mutch5,6,
  11. Matthew A Powell5,6,
  12. Julie K Schwarz1,5,
  13. Perry W Grigsby1,2,5,
  14. Delphine L Chen2,7 and
  15. Stephanie Markovina1,5
  1. 1 Department of Radiation Oncology, Washington University in Saint Louis, Saint Louis, Missouri, USA
  2. 2 Mallinckrodt Institute of Radiology, Washington University in Saint Louis, Saint Louis, Missouri, USA
  3. 3 Division of Biomedical Statistics and Informatics, Mayo Clinic Arizona, Scottsdale, Arizona, USA
  4. 4 Department of Radiation Oncology, The George Washington University School of Medicine and Health Sciences, Washington, District of Columbia, USA
  5. 5 Alvin J Siteman Cancer Center, Washington University in Saint Louis, St. Louis, Missouri, USA
  6. 6 Division of Gynecology Oncology, Department of Obstetrics and Gynecology, Washington University in Saint Louis, St. Louis, Missouri, USA
  7. 7 Department of Radiology, University of Washington, Seattle Cancer Care Alliance, Seattle, Washington, USA
  1. Correspondence to Dr Stephanie Markovina; smarkovina{at}wustl.edu

Abstract

Introduction Metastatic involvement of groin nodes can alter radiation therapy planning for pelvic tumors. 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) can identify nodal metastases; however, interpretation of PET/CT-positive nodes can be complicated by non-malignant processes. We evaluated quantitative metrics as methods to identify groin metastases in patients with pelvic tumors by comparison with standard subjective interpretive criteria, with pathology as the reference standard.

Methods We retrospectively identified patients with vulvar, vaginal, or anal cancers who underwent 18F-FDG PET/CT before pathologic evaluation of groin nodes between 2007 and 2017. Because patho-radiologic correlation was not possible for every node, one index node identified on imaging was selected for each groin. For each index node, standardized uptake value measurements, total lesion glycolysis, metabolic tumor volume, CT-based volume, and short and long axes were measured. Multivariate logistic regression was used to identify metrics predictive for pathologically positive groins and generate a probabilistic model. Area under the receiver-operating characteristic curves (AUCs) for the model were compared with clinical interpretation from the diagnostic report via a Wald’s χ2 test.

Results Of 55 patients identified for analysis, 75 groins had pathologic evaluation resulting in 75 index groin nodes for analysis with 35 groins pathologically positive for malignancy. Logistic regression identified mean standardized-uptake-value (50% threshold) and short-axis length as the most predictive imaging metrics for metastatic nodal involvement. The probabilistic model performed better at predicting pathologic involvement compared with standard clinical interpretation on analysis (AUC 0.91, 95% CI 0.84 to 0.97 vs 0.80, 95% CI 0.71 to 0.89; p<0.01).

Discussion Accuracy of 18F-FDG PET/CT for detecting groin nodal metastases in patients with pelvic tumors may be improved with the use of quantitative metrics. Improving prediction of nodal metastases can aid with appropriate selection of patients for pathologic node evaluation and guide radiation volumes and doses.

  • vulvar and vaginal cancer
  • lymph nodes
  • radiation

https://doi.org/10.1136/ijgc-2020-001363

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    HIGHLIGHTS

    • Index node SUVmean at 50% threshold and short-axis length are significant, independent predictors for pathologically involved groin nodes.

    • Probabilistic model generated from imaging metrics improved prediction for pathologically involved nodes over standard clinical interpretation of 18F-FDG PET/CT images.

    • Improvement in diagnostic accuracy of groin nodes can be utilized to tailor surgical and radiation therapy decisions.

    Introduction

    Obtaining accurate nodal staging with 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) in pelvic malignancies with a propensity for inguinal-femoral (groin) lymph node metastases is more difficult than in other nodal regions due to common local inflammatory or infectious processes that can confound clinical interpretation.1–3 Histopathologic confirmation of nodal status is inconsistently pursued. Surgical candidates often receive nodal dissections to confirm pathologic involvement of groin nodes and guide adjuvant therapy. However, for inoperable patients or for pelvic malignancies such as anal cancer, which are treated with definitive radiation therapy, pathologic assessment is inconsistently performed. Frequently, treatment decisions for radiation therapy volumes and dose are based on clinical exam and clinical interpretation of imaging studies. Standard clinical interpretation of PET/CT has superior negative predictive value compared with clinical examination or CT alone,4–6 but accuracy is lower in the groin nodes than for other nodal regions, and there are currently no accepted guidelines for interpretation of groin lymph nodes like there are for head and neck lymph nodes, for instance.7

    Increasing diagnostic accuracy for groin node metastases could improve surgical and radiation treatment planning by increasing confidence to forego surgical dissection or to employ elective radiation doses. Studies evaluating qualitative interpretations of groin nodes in pelvic malignancies either lacked histopathologic correlation or used PET only techniques.4–6 The prospective GrosNaPET study found PET/CT useful to identify appropriate patients with vulvar cancer for sentinel lymph node biopsy, but the positive predictive value was low at only 38%.8 Other investigators have evaluated dual time-point PET/CT (defined by image acquisition at two time points: 1 and 3 hours after FDG injection) for refining vulvar cancer nodal staging, but reported no significant benefit to qualitative or quantitative interpretations.2 Quantitative metrics in the dual time-point study were limited to maximum standardized-uptake-value and retention index (proportional change in maximum standardized-uptake-value over the two time-points of acquisition). Volumetric parameters such as metabolic tumor volume and total lesion glycolysis of primary tumors are associated with prognostic outcomes for patients with pelvic malignancies, but, to our knowledge, no studies have evaluated volumetric parameters of groin nodes to improve detection of nodal metastases.9–12

    The purpose of our study was to determine if unbiased quantitative PET/CT parameters of groin nodes could improve identification of metastatic groins in patients with vulvar, distal vaginal, and anal cancers as compared with standard clinical-radiologic interpretation.

    Methods

    Study Design and Patient Selection

    This was a single institution, retrospective, cohort study. The study was approved by the Institutional Review Board and requirement for written informed consent was waived. Eligible patients were treated for biopsy-proven vulvar cancer, distal vaginal cancer, or anal cancer from 2007 through 2017 with squamous cell carcinoma histology, and had pathologic assessment of one or both groin regions. Patients presenting with both de novo primary and recurrent disease, prior to treatment of the recurrence, were included in this study. Pathologic evaluation included lymph node dissection, nodal excision, or biopsy. A PET/CT within 1 month of pathologic nodal evaluation with no intervening therapeutic intervention was also required for inclusion.

    Imaging and Quantitative Metrics

    All PET/CT images were obtained using two scanners: Biograph 40 TruePoint or Biograph mCT (Siemens Healthcare; Erlangen, Germany). All patients fasted for at least 4 hours before FDG injection. The studies were performed according to the institutional standard clinical protocol in which images were obtained from the skull base to the mid thighs with arms out of the field-of-view. Non-contrast-enhanced CT images were acquired before PET data acquisition. PET emission data were acquired sequentially with 2–4 min of acquisition per bed position. Data were reconstructed with and without attenuation correction using the 3D-ordered-subset-expectation-maximization reconstruction. The Biograph 40 TruePoint reconstruction parameters included a 168×168 matrix (4.07 mm pixel size) with two iterations, 21 subsets, and a 2 mm Gaussian post-reconstruction filter. The Biograph mCT reconstruction parameters included a 200×200 matrix (4.07 mm pixel size), two iterations, 21 subsets, and a 1 mm Gaussian post-reconstruction filter. There were no changes to reconstruction parameters during the study period.

    Metabolic parameters for nodes were measured using the MIM software package (MIM Software, Inc, Cleveland, Ohio). Primary tumors were not assessed in this nodal staging-focused analysis. PET images were quantitatively analyzed by measuring the metabolic parameters for all radiographically visible groin nodes on the CT images (including non-enlarged nodes and nodes with minimal FDG uptake). Nodes were examined with a sphere-shaped volume of interest that included the entire lesion in the axial, sagittal, and coronal planes of PET images. Three methods were used to create the volumes of interest: maximum standardized-uptake-value of 2.5 (fixed threshold), 42% of the maximum standardized-uptake-value, and 50% of the maximum standardized-uptake-value.10 11 13

    Maximum standardized-uptake-value was determined as the maximal pixel value in the volume of interest. Peak standardized-uptake-value was the highest average standardized-uptake-value within a four contiguous-voxel cluster within the volume and was normalized to lean body mass using the Janmahasatian formula.14 15 Mean standardized-uptake-value was defined as the average standardized-uptake-value within the volume, determined for each of the three threshold volumes of interest. Metabolic tumor volume was obtained by the same volume of interest drawn for mean standardized-uptake-value. Total lesion glycolysis was calculated by multiplying the nodal metabolic tumor volume with its mean standardized-uptake-value. The node volume, length of maximal diameter, and perpendicular short diameter were obtained from the CT portion of the studies.

    Statistical Analysis

    Right and left groin nodal regions were considered individually, such that each patient could have up to two groins evaluated. Direct correlation of individual nodes between pathology and imaging reports was not possible; therefore, a per-groin analysis was performed. As there is a maximum of two observations per patient, intra-subject correlation is not estimable, and therefore each groin was treated as an independent observation. For the purposes of this analysis, the groin node with the highest maximum standardized-uptake-value was identified as the index node for each groin, as has been previously described,2 and all associated imaging metrics for this index node were used for statistical analysis of that groin region. For groins without radiographically visible nodes, imaging metrics were coded as follows: maximum standardized-uptake-value, peak standardized-uptake-value, mean standardized-uptake-value at 42% threshold, and mean standardized-uptake-value at 50% threshold were assigned a value of 1 while all other metrics were assigned 0.

    Standard clinical interpretation of the PET/CT for each groin was classified as radiographically positive, negative, or indeterminate from the diagnostic radiology report. Each groin was characterized as pathologically positive or negative based on pathology reports. Wilcoxon-Rank-Sum tests for continuous variables were performed to compare metrics between pathologically positive and negative groins. Univariate and sequential stepwise multivariate logistic regression modeling techniques were used to identify metrics significantly associated with pathologically positive groins and to generate a probabilistic model based on model fit statistics and stepwise selection criteria, with eligibility at each step set at a significance value of p=0.05. To calculate model-based sensitivity, specificity, positive predictive value, and negative predictive value, the final regression model was employed to calculate the probability of positive pathologic status for each groin. Groins with predicted probability >50% were assigned a positive status; otherwise, a negative status was assigned. Area under the curve (AUC) of the receiver operating characteristic curves for both the model and the standard clinical interpretation were compared via a Wald’s χ2 test. This analysis was performed twice to account for groins characterized as indeterminate in the standard clinical interpretation; first with indeterminate groins assigned as radiographically negative, and then as radiographically positive. Statistical significance was defined as p<0.05, and all testing was two-sided. Statistical analyses were performed using SAS v9.4 (SAS Institute Inc, Cary, NC).

    Results

    A total of 55 consecutive patients (37 patients with vulvar cancer, four with vaginal cancer, and 14 with anal cancer) were identified. Nineteen patients had pathologic evaluation of bilateral groins and 36 had evaluation of one groin resulting in a total of 75 groins evaluated (one patient had the same groin pathologically evaluated for two separate recurrences). Eleven patients (20%) with recurrent disease were included in our cohort and accounted for 13 groins (17%). Five patients had pathologic evidence of bilateral groin metastatic disease, and 25 patients (45%) had unilateral pathologically positive groins. Additional patient and imaging details are presented in Table 1.

    View this table:
    Table 1

    Baseline patient and imaging characteristics

    Each groin was pathologically evaluated via needle biopsy (n=24, 32%), excisional biopsy (n=20, 27%), or groin dissection (n=31, 41%) (Figure 1). Median time between imaging and pathologic evaluation was 8 days. Dissection was performed in 31% and 50% of pathologically positive and negative groins, respectively. The method of pathologic evaluation was not significantly different between pathologically positive versus negative groins (p=0.275). Online supplementary table 1 shows standard clinical interpretation of PET/CT images compared with pathologic groin status. Clinical interpretation classified five groins as indeterminate, four of which were found to be pathologically negative. Sensitivity, specificity, positive predictive value, and negative predictive value of clinical interpretation with indeterminate groins considered positive and negative are shown in Table 2. All evaluated quantitative imaging metrics for the index node were significantly different between pathologically positive and pathologically negative groins (Online supplementary table 2). Maximum standardized-uptake-value was higher for the index lymph node in pathologically positive groins compared with pathologically negative groins (p<0.001). Short-axis length from the CT images ranged from 0.4–5.4 cm in positive groins and 0–2.0 cm in negative groins (p<0.001).

    Supplemental material

    Figure 1
    Figure 1

    Methods of pathologic groin evaluation by laterality and tumor presentation.

    View this table:
    Table 2

    Diagnostic performance of metric model and standard clinical interpretation

    On univariate logistic regression, all metrics were significantly associated with pathologically positive groins. Multivariate logistic regression identified two independent factors associated with pathologically positive groins: mean standardized-uptake-value at 50% threshold (OR 1.7, 95% CI 1.1 to 2.4) and short axis length (OR 6.2, 95% CI 1.5 to 25.8) (Table 3). These metrics were used to generate a probabilistic model which was compared with standard clinical interpretation. The following equation was generated from the logistic regression model to calculate the probability of positive pathologic groins:

    View this table:
    Table 3

    Univariate and multivariate logistic regression for metastatic groin nodes

    Embedded Image

    Model-based prediction of groin status versus pathologic groin status was compared in Online supplementary table 1. The model (AUC 0.91, 95% CI 0.84 to 0.97) improved prediction of pathologically positive groins as compared with standard clinical interpretation (indeterminate groins grouped as negative: AUC 0.80, 95% CI 0.71 to 0.89, p<0.01 (Figure 2A); indeterminate groins grouped as positive: AUC 0.76, 95% CI 0.67 to 0.86, p<0.01 (Figure 2B)). Model-based sensitivity, specificity, positive predictive value, and negative predictive value are reported in Table 2.

    Figure 2
    Figure 2

    (A) Area under the curve (AUC) analysis of receiver operating characteristic (ROC) curves for metric-based model and standard clinical interpretation (indeterminate groins grouped as negative). (B) AUC analysis of ROC curves for metric-based model and standard clinical interpretation (indeterminate groins grouped as positive).

    Discussion

    We have shown that quantitative imaging metrics from diagnostic PET/CT in pelvic malignancies can augment clinical interpretation with a model incorporating mean standardized-uptake-value at 50% threshold and CT short-axis length of the index lymph node with the highest maximum standardized-uptake-value. Semiquantitative and volumetric analyses of primary tumors on PET/CT have been used to determine prognosis, but studies correlating groin lymph node PET/CT metrics with pathologic nodal involvement were lacking. Collarino and colleagues used metrics from groin lymph nodes to predict pathologic lymph node status in patients with vulvar cancer.2 They noted that using a nodal maximum standardized-uptake-value cut-off did not significantly improve sensitivity or specificity over visual interpretation of imaging. In this study, we evaluated volume-based metrics of metabolic tumor volume and total lesion glycolysis at multiple standardized-uptake-value thresholds along with semiquantitative metrics and CT metrics. Both volume-based metrics demonstrated significant association with pathologic positive groin status on univariate analysis, but neither was independently predictive on multivariate analysis. Other investigators have noted a similar lack of predictive power for metabolic tumor volume and total lesion glycolysis in detecting positive groin nodes.16 This was partially due to volume-based metrics being defined by semiquantitative metrics, and as a result, the multivariate analysis was unable to identify metabolic tumor volume or total lesion glycolysis as independent factors. Additional improvements in diagnostic accuracy through volume-based metrics may be too small for detection in this study.

    Studies have demonstrated that PET/CT with standard clinical interpretation improved detection of abnormal groin lymph nodes in pelvic malignancies. Lamoreaux and colleagues demonstrated that PET identified more abnormal lymph nodes in patients with vaginal cancer as compared with CT alone.4 Cotter et al noted improvement in PET/CT detection of abnormal groin nodes in patients with anal cancer over CT alone; they noted that 25% of patients deemed to have no nodal involvement by CT had PET positive nodes.5 However, neither of these investigations evaluated histopathologic correlation of imaging findings, so true sensitivity and specificity could not be determined. Cohn et al reported on a prospective cohort of patients with vulvar cancer who underwent PET only prior to surgical groin evaluation. PET sensitivity was reported as 67% for detecting groin metastases on a per-groin basis.6 Our study also corroborated imaging interpretation with pathology results. Standard clinical interpretation of PET/CT for groin metastases in our study cohort resulted in a sensitivity of 80% on per-groin analysis. These findings are in keeping with multiple prospective series evaluating PET/CT for vulvar cancer nodal staging with sensitivities ranging from 56–95%.1 2 8 PET/CT appeared beneficial when compared with CT alone for identifying metastatic groin involvement, but further improvement in diagnostic accuracy is necessary.

    Clinical interpretation of PET/CT imaging at our institution by experienced academic nuclear medicine physicians demonstrated high accuracy. Thus, the improvement of diagnostic accuracy through the metric-based probabilistic model was noteworthy, with implications for management of patients with pelvic malignancies. Garganese et al prospectively evaluated the use of preoperative PET/CT to improve selection for sentinel node biopsy in clinically node-negative patients with vulvar cancer who would not qualify for sentinel node biopsy under standard recommendations (patients with large primary lesions, multifocal primary, etc). They reported a high negative predictive value for PET/CT (93%) on a per-groin analysis and determined it safely allowed identification of patients who were appropriate for sentinel node procedures, but false positives remained an issue with 38% positive predictive value.8 Our study model improved the positive predictive value over clinical interpretation from 78% to 88%. Employing such a model could be considered to identify patients who are more likely to benefit from pathologic nodal assessment, and possibly decrease rates of complications.17 PET/CT evaluation may be even more critical in patients who are not candidates for surgical evaluation. Patients with locally advanced pelvic malignancies require definitive radiation doses. Radiation volumes often include groin lymph nodes with the prescribed dose depending on whether the groin is considered involved with disease.18–21 Classification of a groin as indeterminate or misclassification based on imaging can make radiation planning challenging. For instance, false negative assessment of groins in patients diagnosed with vulvar cancer and treated with radiation therapy can result in under-dosing of the groins which can subsequently lead to higher recurrence rates.22 Appropriate triaging of groins is of particular importance in patients with vulvar cancer given the significantly higher radiation dose required to adequately treat disease as compared with patients with other pelvic malignancies such as anal cancer.18 21 23 Our model demonstrated significant improvement in prediction over standard clinical interpretation regardless of whether indeterminate reads were categorized as radiographically positive or negative. The proposed metric model had a similar negative predictive value and false negative rate compared with standard clinical interpretation (Table 2), suggesting that a more sophisticated approach is at least not more likely to result in undertreatment of lymph node regions. No consensus was found regarding how best to analyze indeterminate findings, and as a result we performed our analysis by considering indeterminate reads as both positive and negative. This highlighted the robustness of the model and showed that metric-based PET/CT evaluation could improve patient selection for changes in radiation doses and volumes without need for pathologic confirmation. Given the significant role of PET/CT in radiation treatment planning, this could positively impact clinical care.

    Limitations of this study were inherent to its retrospective design. A relatively small patient cohort, heterogenous disease presentation, use of varied methods of pathologic groin evaluation, and differences in scanners and reconstruction techniques used to generate the PET images were additional limitations. Less than half (41%) of groins evaluated in this study underwent dissection which may bias results due to incomplete pathologic assessment from biopsies or node excisions.24 The study included patients with a diagnosis of vulvar, distal vaginal, or anal cancer due to similarities in histology and predilection for groin node metastases, but this does not allow for disease-specific differences. In fact, all patients with anal cancer in this study underwent pathologic evaluation with biopsy which may decrease applicability of the predictive model in this group. In addition, PET/CT interpretation may be more difficult in patients with recurrent disease due to previous treatment-related changes which may obscure small recurrences. Finally, variations in uptake time after FDG injection, the scanner used, and reconstruction techniques are known to affect PET metrics, particularly the maximum standardized-uptake-value. However, all of these limitations would likely decrease the likelihood of finding positive relationships between positive and negative groins. Therefore, the fact that significant differences were identified suggested that our results could lead to improvements in nodal staging in a clinically relevant setting. The generalizability of this model would require an additional analyses of PET/CT imaging outside of what was performed during standard clinical interpretation, though developments in software automation would facilitate such analyses. An outcomes-based analysis is needed to determine whether an imaging metric-based model can be used to guide radiation therapy.

    This study suggested that quantitative metrics improved diagnostic performance of PET/CT in identifying metastatic groins from pelvic malignancies. Although the model presented is not generalizable to all centers, the methodology used in this study can be implemented at institutions with collaborative oncology and radiology groups. As PET/CT is commonplace in the workup of pelvic malignancies, a validated model can potentially improve diagnostic accuracy which can allow clinicians to identify patients who most benefit from surgical groin procedures as well as choose the appropriate radiation dose targeted to the groin to optimize recurrence risk while balancing treatment-related toxicity.

    References

    1. ↵
      2. Lin G ,
      3. Chen C-Y ,
      4. Liu F-Y , et al
      . Computed tomography, magnetic resonance imaging and FDG positron emission tomography in the management of vulvar malignancies. Eur Radiol 2015;25:1267–78.doi:10.1007/s00330-014-3530-1 pmid:http://www.ncbi.nlm.nih.gov/pubmed/25477274
      OpenUrlPubMed
    2. ↵
      2. Collarino A ,
      3. Garganese G ,
      4. Valdés Olmos RA , et al
      . Evaluation of dual-timepoint 18F-FDG PET/CT imaging for lymph node staging in vulvar cancer. J Nucl Med 2017;58:1913–8.doi:10.2967/jnumed.117.194332 pmid:28546331
      OpenUrlAbstract/FREE Full Text
    3. ↵
      2. Perry LTJ ,
      3. Guralp O ,
      4. Al-Niaimi A , et al
      . False positive PET-CT scan and clinical examination in a patient with locally advanced vulvar cancer. Gynecol Oncol Reports 2013.
    4. ↵
      2. Lamoreaux WT ,
      3. Grigsby PW ,
      4. Dehdashti F , et al
      . FDG-PET evaluation of vaginal carcinoma. Int J Radiat Oncol Biol Phys 2005;62:733–7.doi:10.1016/j.ijrobp.2004.12.011 pmid:15936553
      OpenUrlPubMed
    5. ↵
      2. Cotter SE ,
      3. Grigsby PW ,
      4. Siegel BA , et al
      . FDG-PET/CT in the evaluation of anal carcinoma. Int J Radiat Oncol Biol Phys 2006;65:720–5.doi:10.1016/j.ijrobp.2006.01.009 pmid:http://www.ncbi.nlm.nih.gov/pubmed/16626889
      OpenUrlCrossRefPubMed
    6. ↵
      2. Cohn DE ,
      3. Dehdashti F ,
      4. Gibb RK , et al
      . Prospective evaluation of positron emission tomography for the detection of groin node metastases from vulvar cancer. Gynecol Oncol 2002;85:179–84.doi:10.1006/gyno.2002.6605 pmid:http://www.ncbi.nlm.nih.gov/pubmed/11925141
      OpenUrlCrossRefPubMed
    7. ↵
      2. Aiken AH ,
      3. Rath TJ ,
      4. Anzai Y , et al
      . ACR Neck Imaging Reporting and Data Systems (NI-RADS): a White Paper of the ACR NI-RADS Committee. J Am Coll Radiol 2018;15:1097–108.doi:10.1016/j.jacr.2018.05.006 pmid:29983244
      OpenUrlCrossRefPubMed
    8. ↵
      2. Garganese G ,
      3. Collarino A ,
      4. Fragomeni SM , et al
      . Groin sentinel node biopsy and 18F-FDG PET/CT-supported preoperative lymph node assessment in cN0 patients with vulvar cancer currently unfit for minimally invasive inguinal surgery: the GroSNaPET study. Eur J Surg Oncol 2017;43:1776–83.doi:10.1016/j.ejso.2017.06.018 pmid:28751058
      OpenUrlPubMed
    9. ↵
      2. Miccò M ,
      3. Vargas HA ,
      4. Burger IA , et al
      . Combined pre-treatment MRI and 18F-FDG PET/CT parameters as prognostic biomarkers in patients with cervical cancer. Eur J Radiol 2014;83:1169–76.doi:10.1016/j.ejrad.2014.03.024 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24767630
      OpenUrlPubMed
    10. ↵
      2. Burger IA ,
      3. Vargas HA ,
      4. Donati OF , et al
      . The value of 18F-FDG PET/CT in recurrent gynecologic malignancies prior to pelvic exenteration. Gynecol Oncol 2013;129:586–92.doi:10.1016/j.ygyno.2013.01.017 pmid:http://www.ncbi.nlm.nih.gov/pubmed/23369941
      OpenUrlPubMed
    11. ↵
      2. Gauthé M ,
      3. Richard-Molard M ,
      4. Fayard J , et al
      . Prognostic impact of tumour burden assessed by metabolic tumour volume on FDG PET/CT in anal canal cancer. Eur J Nucl Med Mol Imaging 2017;44:63–70.doi:10.1007/s00259-016-3475-5 pmid:http://www.ncbi.nlm.nih.gov/pubmed/27503193
      OpenUrlPubMed
    12. ↵
      2. Bazan JG ,
      3. Koong AC ,
      4. Kapp DS , et al
      . Metabolic tumor volume predicts disease progression and survival in patients with squamous cell carcinoma of the anal canal. J Nucl Med . 2013;54:27–32. [Internet].doi:10.2967/jnumed.112.109470 pmid:http://www.ncbi.nlm.nih.gov/pubmed/23236018
      OpenUrlAbstract/FREE Full Text
    13. ↵
      2. Erdi YE ,
      3. Mawlawi O ,
      4. Larson SM , et al
      . Segmentation of lung lesion volume by adaptive positron emission tomography image thresholding. Cancer 1997;80:2505–9.doi:10.1002/(SICI)1097-0142(19971215)80:12+<2505::AID-CNCR24>3.0.CO;2-F pmid:http://www.ncbi.nlm.nih.gov/pubmed/9406703
      OpenUrlCrossRefPubMed
    14. ↵
      2. Janmahasatian S ,
      3. Duffull SB ,
      4. Ash S , et al
      . Quantification of lean bodyweight. Clin Pharmacokinet 2005;44:1051–65.doi:10.2165/00003088-200544100-00004 pmid:http://www.ncbi.nlm.nih.gov/pubmed/16176118
      OpenUrlCrossRefPubMedWeb of Science
    15. ↵
      2. Tahari AK ,
      3. Chien D ,
      4. Azadi JR , et al
      . Optimum lean body formulation for correction of standardized uptake value in PET imaging. J Nucl Med 2014;55:1481–4.doi:10.2967/jnumed.113.136986 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24963129
      OpenUrlAbstract/FREE Full Text
    16. ↵
      2. Oldan JD ,
      3. Sullivan SA
      . Positron emission tomography-computed tomography for inguinal nodes in vulvar cancer. World J Nucl Med . 2018;17:139–44. [Internet].doi:10.4103/wjnm.WJNM_32_17 pmid:http://www.ncbi.nlm.nih.gov/pubmed/30034276
      OpenUrlPubMed
    17. ↵
      2. Hinten F ,
      3. van den Einden LCG ,
      4. Hendriks JCM , et al
      . Risk factors for short- and long-term complications after groin surgery in vulvar cancer. Br J Cancer . 2011;105:1279–87. [Internet].doi:10.1038/bjc.2011.407 pmid:http://www.ncbi.nlm.nih.gov/pubmed/21970884
      OpenUrlCrossRefPubMed
    18. ↵
      2. Gaffney DK ,
      3. King B ,
      4. Viswanathan AN , et al
      . Consensus recommendations for radiation therapy contouring and treatment of vulvar carcinoma. Int J Radiat Oncol Biol Phys 2016;95:1191–200.doi:10.1016/j.ijrobp.2016.02.043 pmid:http://www.ncbi.nlm.nih.gov/pubmed/27130794
      OpenUrlPubMed
    19. ↵
      2. Greenwalt JC ,
      3. Amdur RJ ,
      4. Morris CG , et al
      . Outcomes of definitive radiation therapy for primary vaginal carcinoma. Am J Clin Oncol 2015;38:583–7.doi:10.1097/COC.0000000000000002 pmid:http://www.ncbi.nlm.nih.gov/pubmed/24136141
      OpenUrlPubMed
    20. ↵
      2. Perez CA ,
      3. Grigsby PW ,
      4. Garipagaoglu M , et al
      . Factors affecting long-term outcome of irradiation in carcinoma of the vagina. Int J Radiat Oncol Biol Phys 1999;44:37–45.doi:10.1016/S0360-3016(98)00530-6 pmid:http://www.ncbi.nlm.nih.gov/pubmed/10219792
      OpenUrlCrossRefPubMed
    21. ↵
      2. Kachnic LA ,
      3. Winter K ,
      4. Myerson RJ , et al
      . RTOG 0529: a phase 2 evaluation of dose-painted intensity modulated radiation therapy in combination with 5-fluorouracil and mitomycin-C for the reduction of acute morbidity in carcinoma of the anal canal. Int J Radiat Oncol Biol Phys 2013;86:27–33.doi:10.1016/j.ijrobp.2012.09.023 pmid:http://www.ncbi.nlm.nih.gov/pubmed/23154075
      OpenUrlCrossRefPubMed
    22. ↵
      2. Koh WJ ,
      3. Chiu M ,
      4. Stelzer KJ , et al
      . Femoral vessel depth and the implications for groin node radiation. Int J Radiat Oncol Biol Phys 1993;27:969–74.doi:10.1016/0360-3016(93)90476-C pmid:http://www.ncbi.nlm.nih.gov/pubmed/8244831
      OpenUrlPubMed
    23. ↵
      2. Rao YJ ,
      3. Chundury A ,
      4. Schwarz JK , et al
      . Intensity modulated radiation therapy for squamous cell carcinoma of the vulva: treatment technique and outcomes. Adv Radiat Oncol 2017;2:148–58.doi:10.1016/j.adro.2017.02.006 pmid:http://www.ncbi.nlm.nih.gov/pubmed/28740926
      OpenUrlPubMed
    24. ↵
      2. Levenback CF ,
      3. Ali S ,
      4. Coleman RL , et al
      . Lymphatic mapping and sentinel lymph node biopsy in women with squamous cell carcinoma of the vulva: a Gynecologic Oncology Group study. J Clin Oncol . 2012;30:3786–91. [Internet].doi:10.1200/JCO.2011.41.2528 pmid:22753905
      OpenUrlAbstract/FREE Full Text

    Footnotes

    • Contributors SR participated in the design of the study, data acquisition, data analysis and drafted the manuscript. DF participated in the data acquisition and drafted the manuscript. TAD and LW performed the statistical analyses and drafted the manuscript. MG, CYH, YJR participated in data acquisition and drafted the manuscript. BAS, FD, DGM, MAP, JKS and PWG participated in data analysis and revised the manuscript. DLC and SM participated in the design of the study, data analysis and revised the manuscript. All authors read and approved the final manuscript.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests BAS has no relevant disclosures to this work, but reports serving in a consultant role for Curium Pharma, ImagingAB, Inc, and Progenics Pharmaceuticals, Inc, serving as advisory board member for GE Healthcare, and his spouse has received lecture honoraria from Siemens. DGM has no relevant disclosures to this work but reports speaking for Clovis and AstraZeneca. MAP has no relevant disclosures to this work but does report relationship with Merck, AstraZeneca, Tesaro, and Clovis Oncology. SM has no relevant disclosures but is supported by NIH K08 CA237822. The contents of this manuscript are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.

    • Patient consent for publication Not required.

    • Ethics approval This study was approved through the institutional review board with waived informed consent (HRPO#201707050).

    • Provenance and peer review Not commissioned; externally peer reviewed.

    • Data availability statement Data are available upon reasonable request. The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.